AMERICAN ASSOCIATION OF ELECTRODIAGNOSTIC MEDICINE REVIEW: QUANTITATIVE SENSORY TESTING EQUIPMENT AND REPRODUCIBILITY STUDIES

Peter Siao Tick Chong, MD; Didier P. Cros, MD

INTRODUCTION
This review is provided as a service to the membership of the American Association of Electrodiagnostic Medicine (AAEM). This paper discusses quantitative sensory testing (QST) equipment using the following sensory modalities: 1) light touch, 2) vibration, 3) thermal, and 4) pain. First, is a summary of the different QST equipment and the equipments specifications as reported in the literature. Second, is a table that summarizes the different reproducibility studies on vibration, thermal, and heatpain thresholds that have been reported since 1981. Other sensory testing devices such as the current perception test, tactile circumferential discriminator, and two-point esthesiometer are not included in the table. This review was based on searches of MEDLINE, and references from relevant articles published between 1966 and 2001. The following search terms were used quantitative sensory testing, QST, sensory threshold, thermal threshold, temperature sense, vibration, and vibration threshold. The search for literature included only articles written in English. No clinical tests or trials were performed by the AAEM or the authors of this review. Neither the AAEM nor the authors of this report reviewed any product literature regarding the pieces of equipment included in the paper nor did the AAEM review the specific equipment. The information included about the equipment and the opinions expressed regarding the reproducibility of each type of equipment are those of the author(s) of the papers cited. The conclusions are not those of the AAEM or the authors of this review. This paper was not created with the intent that it be used as a basis for reimbursement decisions. A literature review of QST can be found in Muscle & Nerve in the May 2004 issue, volume 29, pages 734-747 or on the Muscle & Nerve website at: http://www3.interscience.wiley.com/cgibin/fulltext/108066505/HTMLSTART.

examiner applies the monofilaments on the skin at a right angle, and pressure on the rod is increased slowly until the monofilament buckles or bends. The examiner then asks the patient whether a sensation of pressure was felt. The length and diameter of the monofilaments are standardized and the pressures required to bend the monofilaments are predictable. This device is now called the WeinsteinSemmes pressure aesthesiometer.28 The markings on the rod (1.65 to 6.65) represent the logarithm of 10 times the bowing force in milligrams. The higher number represents a thicker monofilament, which requires a higher pressure to buckle.8 CASE III system This system was developed by Dyck and colleagues to assess cutaneous light touch perception. The apparatus consists of a stylus tip with a diameter of 0.64 mm that is attached to a galvanometer. The stylus tip rests on the skin at a constant load to allow for accommodation. During stimulation, a known force is applied to indent the skin.22 The tactile stimuli are applied to a matrix of 9 points 1 mm apart in a 3 x 3 array and the threshold is estimated using the forced-choice algorithm. The stimulus intensity is given in 21 levels with forces that range from 2.9 mg to 12,000 mg. A forced-choice response paradigm is used to obtain the threshold at each point. The mean threshold for the nine sites is also calculated. The most recent CASE IV system no longer includes this apparatus because the galvanometer motor became unavailable and vibration perception testing may be used to assess the large myelinated sensory fibers.11

Vibration Bio-thesiometer (Bio-Medical Instruments, Newbury, OH) This is a hand-held electromagnetic vibrator with a stimulating probe (12-mm diameter) that vibrates at 100 Hz. It has been in use since 1957. The stimulating probe is placed on the site to be tested, usually the big toe or the finger, and it rests on its own weight (300 g).
The intensity of vibration in increased or decreased by changing the voltage to the stimulator. The vibration threshold is determined by the method of limits and expressed on an arbitrary scale in volts. The operator manually increases the applied voltage until the subject feels the vibration. The voltage is then manually decreased until the sensation disappears.35,23 Being a hand-held instrument has the following limitations: (1) the movement of the stimulator is in more than one plane; (2) the static load on the probe may affect the amplitude of the vibration; and (3) the rate of increase in stimulation intensity is manually controlled and therefore not constant.35 Probably the most

EQUIPMENT
Light Touch von Frey Hairs In 1898, von Frey introduced this method of measuring Semmes and pressure perception with horsehair.28 Weinstein utilized a set of nylon monofilaments with varying diameter and stiffness or bending pressures. The monofilaments are attached to rods at right angles. The

important limitation of this device is the effect of tissue damping on the actual amplitude of vibration. Goldberg and Lindblom showed that the actual vibration amplitude varies depending on tissue consistency of the stimulation site and that different tissues have different damping effects on the vibrator.21 Vibrameter (Somedic AB, Sweden) This instrument is a modification improvement of the BioThesiometer. Instead of using the applied voltage as the measure of stimulus intensity, an accelerometer measures the level of vibration in micrometers. This modification overcomes the problem of varying vibration amplitude secondary to different tissue consistency. The application pressure of this hand-held device on the stimulation site may be held constant with the aid of a load-weighing display. The stimulator generates a sinusoidal vibration at a constant 100-Hz frequency.21 The stimulus intensity or the vibration amplitude is adjusted manually. CASE IV vibration stimulator This device is part of the CASE system that was first introduced by Dyck in 1978.14 It is an automated system that includes a personal computer and software. The system generates the stimuli, cues the patient, records the subjects responses, and determines the sensory threshold according to selected algorithms (WR Medical Electronics Co., Stillwater, MN). Since 1978, the system has undergone modifications of stimulus characteristics, number of stimulus levels, and techniques of finding the sensory threshold. The vibration stimulator is a galvanometer that is mounted on a balance arm to allow the stimulating probe (9 mm diameter) to rest on the base of the nail of the index finger or big toe at a force of 30 grams. The stimulator generates vibration at 125 Hz (previously 250 Hz). The amplitude is controlled by the computer to vary from 0.1 to 576 micrometers in 25 (previously 21 under CASE III) levels or just noticeable difference (JND).22 A cue device informs the subject of the stimuli and the patient responds by pressing on the button of a response box. Vibratron II (Sensortek Inc., Clifton, NJ) This device consists of two identical stimulating posts (1.4-cm diameter) and a controller unit. Only one post vibrates at any one time. The vibration frequency is 120 Hz. The amplitude of vibration varies from 0 to 20 vibration units. The vibration units may be converted to microns of stimulator displacement. The subject is asked to press the digit against each stimulating post for 1 second at a time and identify the probe that is vibrating using the forced-choice protocol. A method of limits protocol has also been used.18,19,20 Vibratory sensory analyzer (Medoc Ltd., Israel) The stimulating probe is mounted on a balance arm that provides a constant weight of 70 grams. The probe faces

up against the pulp of the finger or toe. The contact area is 1.2 cm2 and the vibration frequency is 100 Hz.30 Maxivibrometer (Penn State University, University Park, PA) This device allows sensory loss to be quantilized over a wider range. It measures vibration amplitude from 1 to 1100 mm. The diameter of the stimulating probe is 12.5 mm and the stimulating frequency is 60 Hz. The subject lies prone on the examining table and the device is mounted on a tripod that is not connected to the table to prevent transfer of vibrations through the table.34

Thermal All commercially available thermal stimulators utilize the Peltier principle. The temperature change is made possible by passing a current through two different metals. Cooling occurs at one side of the bimetallic junction while warming occurs on the other side. Kenshalo introduced this method for the study of temperature sense in humans and subhuman species.27
Marstock stimulator (Somedic AB, Sweden) Fruhstorfer, Lindblom, and Schmidt described this device in 1976. It is called the Marstock stimulator because of the collaboration between investigators from the two cities of Marburg and Stockholm.17 The Peltier device and a thermocouple are placed in the thermode and circulating water cools the device. The thermode is placed over the palmar aspect of the thenar region or the lateral aspect of the dorsum of the foot. The subject is instructed to press a button as soon as a warm sensation is perceived. The direction of temperature change reverses to cool when the button is pressed. The temperature is measured with a thermocouple and plotted by a pen recorder. The warm-cold difference limen is then measured from the temperature recording. This original method evaluates warm and cool sensation together. It is a quick way to assess temperature sensation. Moreover, hypoesthetic and hyperesthetic conditions can be evaluated. Subsequent modifications of this method have been described to measure warm and cool thresholds separately.6,29,36 CASE IV thermal testing probe This instrument is part of the CASE IV automated system (WR Medical Electronics., Stillwater, MN). In contrast to the thermode in CASE III, this thermode is now water-cooled to allow a faster rate of temperature change. The stimulating surface that comes in contact with the skin is a ceramic sheet (10 cm2) with high conductance. The thermode consists of two thermoelectric units (TEU 1 and TEU 2) and a thermocouple attached to the ceramic sheet. An aluminum block separates the two TEUs. Another aluminum block is mounted on TEU 2 and water circulates around it to dissipate heat. TEU 1 provides the warm or cold temperature ramps and TEU 2 maintains the aluminum block at skin temperature. The patients skin temperature is used as the baseline. The thermodes temperature is maintained by the computer through the

thermodes temperature ranges from 5C to 50C, and the stimulus intensity is given in 25 levels or JNDs. The rate of temperature change is set at 4C/s. The maximum temperatures that can be attained are the following: 9C x 10 s for cold perception threshold, 45C x 10 s for warm perception threshold, and 49C x 10 s for heat pain perception.11,22 Glasgow system This is an automated system used to measure warm and cool thresholds. The original thermode was watercooled,25,26 but a portable system that did not require water cooling became available as the Triple T (Thermal Threshold Tester, Medelec).5 The stimulating thermode has a surface area of 12.5 cm2. Using a constant rate of temperature change (1C/s), the stimulus intensity is graded by changing the duration of stimulus application. The baseline skin temperature is maintained at 34C. The thermal threshold is defined as the minimal temperature change from the baseline skin temperature that the subject can reliably perceive.5 Middlesex Hospital thermal testing system This is another computer-driven portable system with no water cooling system. The stimulating surface area is 7.5 cm2 for the hand and 26 cm2 for the foot.16 Similar to the Glasgow system, the rate of temperature change is kept constant at 1C/s and the stimulus duration determines the stimuli intensity. The warm and cool thresholds are determined by utilizing the yes-no method to obtain the subjects responses and the staircase algorithm to change the stimulus intensities.16 Thermal sensitivity tester (Sensortek, Inc., Clifton, NJ)

This device consists of two 25 cm2 nickel-coated copper plates connected to different power units and cooled with water. The subject alternately presses on each plate for 2 seconds. The temperature of one plate changes on a series of fixed-step digital controls while the temperature of the other plate is maintained at 25C. The temperature difference between the plates is displayed on a meter.1 Thermal sensory analyzer TSA-2001 (Medoc Ltd., Israel) This is also a microprocessor driven automated system. The thermode size is 50 x 25 mm2 and the temperature range is 0-50C. A computer feedback mechanism controls the temperature of the thermode. The yes-no method is used to obtain subjects responses. The stimulus intensities are adjusted with an initial step of 4C with subsequent steps reduced in half until step size reached 0.2C. Unlike the CASE IV system, no visual or auditory cues are given to signal stimulus onset.36 PATH-tester MPI 100 (PHYWE Systeme GmbH, Germany) This is a microcomputer-driven modified Marstock thermode. The thermode contact surface measures 1.6 x 3.6 cm. The temperature range is 17C to 50C and it is controlled by a feedback circuit.6

REPRODUCIBILITY
The following table summarizes the reproducibility studies on vibration, thermal, and heat-pain thresholds that have reported in the literature since 1981. The conclusions are those of the author(s) of the papers cited and not those of the AAEM or the authors of this article.

REPRODUCIBILITY STUDIES ON QST

REF 34 SUBJECTS 7 diabetics with neuropathy and 7 matched nondiabetic controls. INTERVAL 3 times daily on 2 separate days within 3 weeks MODALITY Vibration EQUIPMENT Maxivibrometer (Penn State University, University Park, PA) Heel probe area: 12.5 mm diameter Frequency: 60 Hz. Amplitude range: 1 to 1100 micrometer METHODOLOGY Method of Limits. As the stimulus intensity increased, the subject pressed the button as soon as vibration was perceived. Computer recorded this as appearance threshold. Then, the stimulus intensity decreased and the subject pressed the button as soon as the vibration was no longer felt (disappearance threshold). The procedure was repeated 3 times so that 6measurements were made. The average of these 6 measurements was considered the threshold. Stimulation site: right hallux. 1 examiner only. a. Warm, cold thresholds: Method of Limits. Skin adaptation temperature 32C, rate of return 1 C/s, average of 4 readings, 6 s interstimulus interval, rate of return 10C/s, average of 3 readings, 10 s interstimulus interval. b. Warm, cold thresholds: Method of Levels. Yes or no response. Initial temperature step of 3C. Subsequent step sizes were increased or decreased at one half the initial intensity. Null stimuli included. Test terminates when step size decreased to 0.1C. Threshold = average of the last yes and no response. c. Vibration: Method of Limits. Linearly increasing train of 4 stimuli. Intensity starting at 0 m amplitude and increased at 0.1 m/s. Threshold = average of 4 consecutive determinations. Method of Limits. Vibration threshold is the average of 3 appearance thresholds and 3 disappearance thresholds. Catch trials with no stimulation or steady stimuli were included. Method of Levels. 2 alternative (spatial) FC. 1 examiner only. 2 identical probes were presented to the subject. The subject is asked to identify which one was actually vibrating. The test began at the highest intensity of 20 vibration units (VU) or at a lower intensity appropriate for the patient. The intensity was decreased by 10% after each correct response and increased by 10% after each wrong response. 5 wrong answers were enough to finish the examination. The lowest and highest value of 10 errors and correct responses were eliminated and the mean of the remaining 8 score was called the VT. Monofilaments applied on both feet at the first toe, medial surface and base of the third metatarsal bone. Threshold was defined as the total number of times the application of the monofilaments was not felt. This could vary between 0 and18. MEASURE OF REPRODUCIBILITY Replication to replication ICC Controls 0.980 Diabetics 0.956 Day to day ICC Controls 0.779 Diabetics 0.940 Wilcoxon signed-ranks test with a Bonferroni correction for multiple comparisons. Hand and foot. p>0.05 CONCLUSION*

Excellent reliability

30

101 NS, age 6-17

2-4 weeks

Warm, Cold Vibration

Medoc TSA-2001 (Medoc, Israel) Thermode area: 3 x 3 cm Temperature range: 0-50C Rate of change: 1C/s Medoc Vibratory Sensory Analyzer (Medoc, Israel) Frequency: 100 Hz Probe area: 1.2 cm2 Constant weight: 70 g

a. No significant session differences b. No significant session differences c. No significant session differences

24

49 children, age 3.3 to 6.8

1 to 58 days (27.322.9)

Vibration

31

23 diabetics with neuropathy

3 times over 21 days with 1 to 14 day intervals

Vibration

Vibrameter (Stockholm, Sweden) Frequency: 120 Hz Probe diameter: 13 mm Constant weight: 650 g Vibraton II (Clifton, NJ) Probe area 1.5 cm
2

2-sided Wilcoxon signed ranks test: p>0.1 2-tailed Spearman rank test: r = 0.590, p<0.01 CV%/ SD left hand 17.69/ 11.28 Right hand 17.61/ 14.92 left toe 20.60/ 16.99 right toe 19.46/ 11.40

High reproducibility

Reproducible

33

64 diabetics

2 to 4 weeks and Same day

Pressure

Semmes Weinstein monofilaments (sizes 4.17, 5.07, 6.10)

13

25 NS

37

72-76 NS, age 20-59

Twice at intervals of more than a day but less than 2 weeks 2 weeks

Heat-Pain

CASE IV (WR Medical Electronics Co. Stillwater, MN)thermode area 10 cm2 rate of change 4C/s. Medoc TSA-2001 (Medoc, Ramat, Yishai, Israel) thermode area 46 x 30 mm2

Non-repeating with null stimuli algorithm (see text for details).

intraobserver test-retest mean (SD) 0.39 (0.35), intrasubject SD 0.16, CV 0.41, CC 0.80 interobserver test-retest mean (SD)0.40 (0.37), intrasubject SD 0.18, CV 0.41, CC 0.77 HP:0.5 %- to76% within 1 stimulus step HP:5.0% to 88% within 1 stimulus step

Sufficiently Reproducible

Sufficiently Reproducible Low variability

Heat-Pain

Method of Limits. Skin adaptation temperature 32C, rate of temperature change 2C/s, average of 3 readings, 20 s interstimulus interval.

ANOVA-based model by Techno-Stat r value Thenar 5.85** Foot 4.47 ** significant intersession bias r value- 95% confidence that 2 measurements on the same patient would differ by less than r.

Poor Sufficient

REF 36

SUBJECTS 72-76 NS, age 20-59

INTERVAL 2 weeks

MODALITY Warm, Cold

EQUIPMENT Medoc TSA-2001 (Medoc, Israel) thermode 50 x 25 mm2 20 Hz feedback, mechanism to maintain linear temperature change

METHODOLOGY a. Method of Limits. Skin adaptation temperature 32C, rate of temperature change 1C/s, cold and warm perception thresholds, average of 3 readings, 6 s interstimulus interval. b. Method of Levels: (Modified from Yarnitsky and Ochoa 1990, 1991), yes or no response, initial step of 40C, reduced by half until step size reached 0.2C. c. Method of Levels: Staircase algorithm (Fowler 1987), yes or no response, initial temperature step of 4C, subsequent steps at 1C, then 0.2C, test terminates after 4 no responses.

MEASURE OF REPRODUCIBILITY ANOVA-based model by Techno-Stat r value a. Thenar cold 1.964** Warm 1.587** Foot cold 3.778 warm 4.298 b. Thenar cold 1.040 warm 0.572 Foot cold 3.016 warm 3.758 c. Thenar cold 1.144 warm 0.720 ** significant intersession bias Total variation coefficient (%) Warm 64.5 Cold 116.6 Vib mm 35.3 Vib toe 41.0 Intra-subject variability (% total variance) Warm 32.8 Cold 15.7 Vib mm 21.0 Vib toe 21.0 30 NS CC (correlation coefficient) vibration toe

CONCLUSION*

Poor Good Good Good Good

32

132 diabetics with moderate neuropathy, age 18-66, mean 45.4

4 weeks

Warm, Cold Vibration

Thermal Testing System (London, UK) Bio-Thesiometer (Bio-Medical Instruments, Newbury, OH)

A multicenter study, FC, ascending and descending intensity ramps. Poor Poor Satisfactory Satisfactory

30 NS, 12 diabetics

Within 1 week

Vibration Warm Cold

Vibratester 100 (Phyne systeme GmbH, D3400, Gottinger, Germany), frequency 100 Hz, amplitude 0-150m Warm, Cold, computer-driven modified Marstock thermodePATH-Tester MPI 100 (Phyne Systeme GmbH, D3400. Gottinger, Germany). thermode size 1.6 x 3.6 cm Ramp rate 1C/s Reference temp 35C Temperature limits 17 to 50C

Vibration: a. Method of Limits. The subject is asked to press a button when vibration is perceived with an increasing ramp (perception threshold) and when vibration disappeared with a decreasing ramp (disappearance threshold). Ramp rate- 0.2:m/s threshold is calculated from the mean of 3 vibration perception thresholds and 3 vibration disappearance thresholds. b. 2 alternative (temporal) FC. Up, down transformed rule, amplitude changes are in steps of 25% of the first stimulus. Threshold- the mean of 5 upper and lower reversal values. c. Titration Method. Stimulus starts with a suprathreshold stimulus (VT + VT1/2) stimulus amplitude is reduced in small steps until it disappears, catch trials inserted at random, threshold- mean of 5 nonperceived stimuli with reversing effect and their preceding perceived stimuli. Warm and Cold: Method of Limits. 7 warm and 7 cold stimuli are applied; the subject presses a button when temperature change is perceived. The 1st stimulus is disregarded and the mean of 6 values represents the threshold.
2 alternative (temporal) FC. Initial reference temperature 34C, rate of change 1C/s, uniform size of steps. Use of varying stimulus duration to change stimulus intensity level with the up-down transformed rule.

39 diabetics without neuropathy, mean age 56.78.6

2 weeks

Warm, Cold

Thermal Threshold Tester (Medelec, Woking, UK)

Limits 0.92 FC 0.96 Titration 0.89 warm 0.77 cold 0.66 90th %tile of absolute day-to-day difference vib toe Limits 0.97 FC 0.75 Titration 1.37 warm 2.3 cold 3.1 Magnitude of day-to-day differences (% of mean value) vib toe Limits 2.9 FC 11.0 Titration 6.0 warm 2.0 cold 9.0 12 diabetics, CC vib toe 0.87 warm 0.92 cold 0.78 Repeatability coefficient is twice the SD. SD is calculated from the mean of all differences between measurement 1 and 2 (after 2 weeks). Normal/abnormal Warm hand 0.19/1.17 Cold hand 0.17/1.01 Warm foot 4.34/--Cold foot 0.60/4.69 --- Insufficient sample size

Good Good Good Good Good

Good Good Good

Acceptable Acceptable Poor/--Acceptable/poor

REF 2

SUBJECTS 49 NS

INTERVAL Vibration 11 months Thermal 7 months

MODALITY Vibration Cold Warm Vibration Cold Warm

EQUIPMENT Bio-Thesiometer (Bio-Medical Instruments, Newbury, OH) Computer controlled thermal stimulator (reference: Fowler 1987)

METHODOLOGY A multicenter study (see text- Fowler 1987)

MEASURE OF REPRODUCIBILITY Malleolus/ big toe 29.7/ 20.8 Male/female Cool 71.0/61.1 Warm 66.8/60.3 Diabetics: Vibration Malleolus/ big toe 6.1-28.2/ 8.7-29.7 Cool 8.8-129.5 Warm 3.2-108.1 Warm >0.8 ICC r1 value Cold >0.9 Vibration >0.9 r1 value- proportion of this variance attributable to variability among subjects ranging from 0 to 1. 0 = all variability is experimental error. 1 = no error. Index finger < 4 JND 75th percentile < 2 JND Great toe < 4 JND 75th percentile < 2 JND At 1 JND/s Index finger 2.1 JND 75th percentile 1.1 JND Great toe <6 JND 75 percentile 2 JND At 4.15m/s Index finger < 2 JND th 75 percentile < 1 JND Great toe 4 JND 75th percentile < 2 JND Retest Reliability coefficient Rtt FC/method of limits Warm threshold: 1st-2nd trial 0.89/0.73 2nd-3rd trial 0.88/0.83 1st to 3rd trial poor Cold threshold: 1st to 2nd trial 0.82/0.71 2nd to 3rd trial 0.78/0.71 1st to 3rd trial poor CV (mean) Normal Subjects Method of Limits Warm 14% Cool 42% 29% test 1 1.75 test 2 1.04 Warm Cold CV(%) NS: Vibration

CONCLUSION*

Reproducible Poor Poor Reproducible Poor Poor High Reproducibility

397 diabetics

Vibration and thermal 1 month.

12

20 diabetics with and without neuropathy

3 to 5 days

Warm, Cold Vibration

CASE III/IV systems*, thermode 10 cm2 CASE III/IV systems,** Stimulus frequency 250 Hz in CASE III, and 120 Hz in CASE IV **Modifications were made in the CASE IV system CASE IV: vibratory transducer, frequency 120 Hz

2 alternative (temporal) FC. Up down transformed rule, algorithm, thermode algorithm for obtaining subjects responses and for determination of thresholds were modified in CASE IV. The 4, 2, 1 stepping algorithm was not used in this study. Thermode was modified in CASE IV. a. 2 alternative (temporal) FC. Up down transformed rule.

10

20 NS

Separate days

Vibration

Accurate and repeatable Accurate and repeatable Good repeatability but overestimated threshold

b. Method of Limits. Bekesy algorithm with null stimuli.

c. Method of Limits. Linear ramp algorithm with and without null stimuli.

55 NS, 28 f , 27 m, age 20-79, mean age 41

3 consecutive days

Cold, Warm

Computer-driven modified Marstock- PATH-Tester MPI 100 (PHYNE Systeme GmbH, Germany) Thermode size 1.6 x 3.6 cm Ramp rate 1C/s Reference temp 35C Temperature limits 17 to 50C

2 alternative (temporal) FC. Up and down transform rule. Initial steps of 0.5C, then steps of 0.1C, after 6 changes in direction, the mean of reversing values represents the threshold. Warm and cold thresholds are measured separately. Method of Limits. 7 warm and 7 cold stimuli are applied; the subject presses a button when temperature change is perceived. The first stimulus is disregarded and the mean of 6 values represents the threshold.

Sufficient reproducibility for 1st-2nd and 2nd3rd trial Vary considerably for1st with the 2nd or the 3rd trial

29

5 NS

3 occasions over 3 to 6 weeks Twice in 1 to 12 weeks

Thermal

Somedic modification of Marstock method Thermal Sensitivity Tester (Sensorteck, Inc., Clifton, NJ)

Method of Limits. Warm and cool thresholds tested separately. 2 alternative (spatial) FC. 1 plate maintained at 30C, up-downtransformed rule, threshold is mean of 6 turnaround points.

131 diabetics

Method of Levels Diabetics Method of Levels Mean SD Thermal 2 thermostimulators 3 x 4 cm 2 alternative (spatial) FC: First thermode is maintained at 5C above or below skin temperature when evaluating warm or cold thresholds respectively. Second thermodes temperature is varied at a random sequence. Initial temperature difference between the 2 thermodes is 10C and is changed according to the up-down transformed rule. Correct answer = CCC, CCWC Wrong answer = CW,CCWW Thermal discrimination threshold is defined as the mean value of the last 6 reversal points. CV

Suitable for clinical use Suitable for clinical use Suitable for clinical us. Large intraindividual variability

71 NS, M:F 29:42, age 21 to 92

Twice on separate days.

1.51C 0.71C 58.4% 56.0%

REF 1

SUBJECTS 10 NS

INTERVAL 10 times at intervals of at least a day. 1 week

MODALITY Thermal

EQUIPMENT Thermal Sensitivity Tester (Sensorteck, Inc., Clifton, NJ), thermode 25 cm2 Modified Bio-thesiometer (BioMedical Instruments, Newbury, OH)

METHODOLOGY 2 alternative (spatial) FC: 1 plate maintained at 25C, initial temperature difference 6C, steps of 10% up and down Vibration perception threshold is the mean of 5.

MEASURE OF REPRODUCIBILITY Mean CV, 19% index finger, 26.6% great toe, CV range 8.3% to 47.1% Pearson CC NS/diabetics 0.82/0.81 Intra-individual variability-change in the mean values of thresholds expressed as a percentage of the first measurement Normal/ diabetics 4.91%/12.85% CC Vibration: NS Hands r = 0.83 p<0.01 Feet r = 0.86 p<0.01 Diabetics Hands r = 0.73 p<0.01 Feet r = 0.91 p<0.01 Thermal: NS Hands r = 0.69 p<0.01 Feet r = 0.79 p<0.01 Diabetics Hands r = 0.91 p<0.01 Feet r = 0.89 p<0.01 Change in mean (of the group) thermal threshold < 5% CV 0% to 6%

CONCLUSION* Reliable

23

25 NS, 60 diabetics (39 with neuropathy, 21 without), M:F 30:30, age 17 to 73, mean age 1.82.1 M:F 33:21, age 21 to 55, mean age 34.3 18 NS, 20 diabetics (vibration) 18 NS, 18 diabetics (thermal)

Vibration

Satisfactory in NS and most diabetics

4 weeks

Vibration

Bio-Thesiometer (Biomedical Instruments, Newberry, OH) Marstock thermostimulator (Somedic, Sweden), thermode 2.5 x 5.0 cm, rate of change 1C/s Glasgow Thermal System Computer driven thermode current calculated every 100 m/s Thermode 2.5 x 5.0 cm, water cooled

Vibration threshold is the mean of 3 trials.

No significant differences

Thermal

Method of Limits (warm-cold limen)

Reproducible

25

106 NS, age 6 to 73 mean age 33.0, SD 17, M:F 45:61 2NS

Short term: 24 hours Long term: 2 to 8 weeks, 17 times daily

Thermal

2 alternative (temporal) FC with null stimuli: 1 person performed all tests. Initial skin temperature 34C to 35C, rate of change 1C/s up down transformed rule uniform size of steps use of varying stimulus duration to change the stimulus intensity level. Site of stimulation: mid-forearm.

Small intraindividual variation

30 NS, age 24 to 91, M:F 11:25 20 diabetics with neuropathy, age 22 to 69, M:F 9:11

1 week

Thermal

2 thermostimulators, Thermode size 3 x 4 cm

15

13 NS, age 25 to 63, mean age 43.7, M:F 5:8 27 patients with neuropathy of various etiologies age 27 to 71, mean age 53.2, M:F 22:5

Short interval Vibration, 5 NS, 4 times within 5 to 10 minutes. Long interval 13 NS, 4 consecutive days in 1 wk. 4 times at intervals of 1, 3, 5 wks. 27 patients, 4 to 5 times in 7 to 13 wks.

Vibration

Hand-held bioThesiometer Vibrameter (MUAB AB, Sweden) 100 Hz Tactile Stimulator (Lindblom) Marstock Stimulator

2 alternative (spatial) FC: First thermode is maintained at skin temperature. Second thermodes temperature is varied at a random sequence. Initial temperature difference between the 2 thermodes is 10C and is changed according to the up-down transformed rule. Correct answer = CCC, CCWC Wrong answer = CW,CCWW Thermal discrimination threshold is defined as the mean value of the last 6 reversal points. Average of at least 2 up and down values. Method of Limits (warm-cold limen)

Intra-individual variability expressed as maximum difference of TDTs between the 1st and second trial. 0.5C for the hands and feet of NS and for the hands of diabetics. Note: 3 diabetics were excluded. 1st and 2nd trial TDTs were as follows: 0.4C to 3.0C 0.6C to 7.5C 0.5C to 4.0C % change from first determination. Short term, vibration only, 8% to 18%

Small intraindividual variation Small intraindividual variability in all NS and in most diabetic patients

Short term: Limited variation Long term: Pronounced variation

Touch

Thermal

Long term, all types of threshold -90% to +256% On average, a change of <60% or >+150% from initial values is required to ensure that a subsequent value will reflect a true change with probability of 95%.

ANOVA = analysis of variance; C = Correct; CC = Correlation Coefficient; CCV = Correlation Coefficient Variation; CV = Coefficient of Variation; FC = Force Choice; HP = Heat Pain; ICC = Interclass Correlation Coefficient; JND = just noticeable difference; M:F = Male:Female; mm = medial malleoli; NS = Normal Subjects; s = second; SD = Standard Deviation; TDT = Thermal Discriminating Thresholds; vib= vibration; VT = Vibration Threshold; VU = Vibration Units; W = Wrong; wks = weeks y = years. ** The conclusion listed in this column is that of the author(s) of the paper cited in the reference. The conclusion is not the opinion of the AAEM.

Disclaimer: No clinical tests or trials were performed by the AAEM or the authors of this review. Neither the AAEM nor the authors of this report reviewed any product literature regarding the pieces of equipment included in the paper nor did the AAEM review the specific equipment. The information included about the equipment and the opinions expressed regarding the reproducibility of each type of equipment are those of the author(s) of the papers cited. The conclusions are not those of the AAEM or the authors of this review. This paper was not created with the intent that it be used as a basis for reimbursement decisions.

REFERENCES
Arezzo JC, Schaumburg HH, Laudadio C. Thermal sensitivity tester. Device for quantitative assessment of thermal sense in diabetic neuropathy. Diabetes 1986;35:590-592. 2. Armstrong FM, Bradbury JE, Ellis SH, Owens DR, Rosen I, Sonksen P, Sundkvist G. A study of peripheral diabetic neuropathy. The application of age-related reference values. Diabet Med 1991;8 Spec No:S94-S99. 3. Bertelsmann FW, Heimans JJ, van Rooy JC, Heine RJ, van der Veen EA. Reproducibility of vibratory perception thresholds in patients with diabetic neuropathy. Diabetes Res 1986;3:463466. 4. Bertelsmann FW, Heimans JJ, Weber EJ, van der Veen EA, Schouten JA. Thermal discrimination thresholds in normal subjects and in patients with diabetic neuropathy. J Neurol Neurosurg Psychiatry 1985;48:686-690. 5. Bravenboer B, van Dam PS, Hop J, vd Steenhoven J, Erkelens DW. Thermal threshold testing for the assessment of small fibre dysfunction: normal values and reproducibility. Diabet Med 1992;9:546-549. 6. Claus D, Hilz MJ, Neundorfer B. Thermal discrimination thresholds: a comparison of different methods. Acta Neurol Scand 1990;81:533-540. 7. Claus D, Mustafa C, Vogel W, Herz M, Neundorfer B. Assessment of diabetic neuropathy: definition of norm and discrimination of abnormal nerve function. Muscle Nerve 1993;16:757768. 8. Dellon AL, Mackinnon SE, Brandt KE. The markings of the Semmes-Weinstein nylon monofilaments. J Hand Surg 1993;18:756-757. 9. Doeland HJ, Nauta JJ, van Zandbergen JB, van der Eerden HA, van Diemen NG, Bertelsmann FW, Heimans JJ. The relationship of cold and warmth cutaneous sensation to age and gender. Muscle Nerve 1989;12:712-715. 10. Dyck PJ, Karnes JL, Gillen DA, OBrien PC, Zimmerman IR, Johnson DM. Comparison of algorithms of testing for use in automated evaluation of sensation. Neurology 1990;40:1607-1613. 11. Dyck PJ, Karnes JL, OBrien PC, Zimmerman IR. Detection thresholds of cutaneous sensation in humans. In: Dyck PJ, Thomas PK, Griffin J, Low PA, Poduslo JF, editors. Peripheral Neuropathy, Vol 1; Philadelphia: WB Saunders; 1993. pp 706-728. 12. Dyck PJ, Kratz KM, Lehman KA, Karnes JL, Melton LJ, OBrien PC, Litchy WJ, Windebank AJ, Smith BE, Low PA, Service FJ, Rizza RA, Zimmerman BR. The Rochester diabetic neuropathy study: design, criteria for types of neuropathy, selection bias, and reproducibility of neuropathic tests. Neurology 1991;41:799-807. 13. Dyck PJ, Zimmerman IR, Johnson DM, Gillen D, Hokanson JL, Karnes JL, Gruener G, OBrien PC. A standard test of heat-pain responses using CASE IV. J Neurol Sci 1996;136:54-63. 14. Dyck PJ, Zimmerman IR, OBrien PC, Ness A, Caskey PE, Karnes J, Bushek W. Introduction of automated systems to evaluate touch-pressure vibration, and thermal cutaneous sensation in man. Ann Neurol 1978;4:502-510. 1.

15. Fagius J, Wahren LK. Variability of sensory threshold determination in clinical use. J Neurol Sci 1981;51:11-27. 16. Fowler CJ, Carroll MB, Burns D, Howe N, Robinson K. A portable system for measuring cutaneous thresholds for warming and cooling. J Neurol Neurosurg Psychiatry 1987;50:1211-1215. 17. Fruhstorfer H, Lindblom U, Schmidt WG. Method for quantitative estimation of thermal thresholds in patients. J Neurol Neurosurg Psychiatry 1976;39:1071-1075. 18. Gerr FE, Letz R. Reliability of a widely used test of peripheral cutaneous vibration sensitivity and a comparison of two testing protocols. Br J Ind Med 1988;45:635-639. 19. Gerr F, Letz R, Harris-Abbott D, Hopkins LC. Sensitivity and specificity of vibrometry for detection of carpal tunnel syndrome. J Occup Environ Med 1995;37(9):1108-1115. 20. Gerr F, Letz R, Hershman D, Farraye J, Simpson D. Comparison of vibrotactile thresholds with physical examination and electrophysiological assessment. Muscle Nerve 1991;14:1059-1066. 21. Goldberg JM, Lindblom U. Standardised method of determining vibratory perception thresholds for diagnosis and screening in neurological investigation. J Neurol Neurosurg Psychiatry 1979;42:793-803. 22. Gruener G, Dyck PJ. Quantitative sensory testing: methodology, applications and future directions. J Clin Neurophysiol 1994;11:568-583. 23. Guy RJ, Clark CA, Malcolm PN, Watkins PJ. Evaluation of thermal and vibration sensation in diabetic neuropathy. Diabetologia 1985;28:131-137. 24. Hilz MJ, Axelrod FB, Hermann K, Haertl U, Duetsch M, Neundrfer B. Normative values of vibratory perception in 530 children, juveniles and adults aged 3-79 years. J Neurol Sci 1998;159:219-225. 25. Jamal GA, Hansen S, Weir AI, Ballantyne JP. An improved automated method for the measurement of thermal thresholds. I. Normal subjects. J Neurol Neurosurg Psychiatry 1985;48:354-360. 26. Jamal GA, Weir AI, Hansen S, Ballantyne JP. An improved automated method for the measurement of thermal thresholds. II. patients with peripheral neuropathy. J Neurol Neurosurg Psychiatry 1985;48:361-366. 27. Kenshalo DR, Bergen DC. A device to measure cutaneous temperature sensitivity in humans and subhuman species. J Appl Physiol 1975;39:1038-1040. 28. Levin S, Pearsall G, Ruderman RJ. Von Freys method of measuring pressure sensibility in the hand: an engineering analysis of the Weinstein-Semmes pressure aesthesiometer. J Hand Surg 1978;3:211-216. 29. Levy D, Abraham R, Reid G. A comparison of two methods for measuring thermal thresholds in diabetic neuropathy. J Neurol Neurosurg Psychiatry 1989;52:1072-1077. 30. Meier PM, Berde CB, DiCanzio J, Zurakowski D, Sethna NF. Quantitative assessment of cutaneous thermal and vibration sensation and thermal pain detection thresholds in healthy children and adolescents. Muscle Nerve 2001;24:1339-1345. 31. Nasseri K, Strijers RL, Dekhuijzen LS, Buster M, Bertelsmann FW. Reproducibility of different methods for diagnosing and monitoring diabetic neuropathy. Electromyogr Clin Neurophysiol 1998;38:295-299. 32. Valensi P, Attali JR, Gagant S. French Group for Research and Study of Diabetic Neuropathy. Reproducibility of parameters for assessment of diabetic neuropathy. Diabet Med 1993;10:933939. 33. Valk GD, de Sonnaville JJ, van Houtum WH, Heine RJ, van Eijk JT, Bouter LM, Bertelsmann FW. The assessment of diabetic polyneuropathy in daily clinical practice: reproducibility and validity of Semmes Weinstein monofilaments examination and clinical neurological examination. Muscle Nerve 1997;20:116-118. 34. van Deursen RW, Sanchez MM, Derr JA, Becker MB, Ulbrecht JS, Cavanagh PR. Vibration perception threshold testing in patients with diabetic neuropathy: ceiling effects and reliability. Diabet Med 2001;18:469-475. 35. Yarnitsky D, Fowler CJ. Quantitative sensory testing. In Binnie CD, Cooper R, Fowler CJ, Mauguiere F, Prior PF, Osselton JW, editors. Clinical neurophysiology- EMG, nerve conduction and evoked potentials. Oxford: Butterworth-Heinemann; 1995. pp 253-270. 36. Yarnitsky D, Sprecher E. Thermal testing: normative data and repeatability for various test algorithms. J Neurol Sci 1994;125:39-45. 37. Yarnitsky D, Sprecher E, Zaslansky R, Hemli JA. Heat pain thresholds. normative data and repeatability. Pain 1995;60:329-332.